1
00:00:00,000 --> 00:00:06,269
Welcome to Gatekeeper Q A. This guide shows a real browser scan, a saved review, a report export,

2
00:00:06,269 --> 00:00:12,539
and a comparison after a fix. We use our own demonstration page. These are captured steps with pauses

3
00:00:12,539 --> 00:00:19,808
shortened, and synthetic narration. Automated checks support human testing; they do not certify accessibility.

4
00:00:19,808 --> 00:00:24,125
Open scan dot gatekeeper Q A dot com. Enter one public page you own or have permission to

5
00:00:24,125 --> 00:00:28,442
test. Our target is slash guide dash demo. The hosted scanner needs a public address. Use the developer

6
00:00:28,442 --> 00:00:32,758
C L I for localhost. This recording is already signed in, so the report can be retained in

7
00:00:32,758 --> 00:00:38,075
the account.

8
00:00:38,075 --> 00:00:43,783
Confirm the permission checkbox, then choose Run accessibility scan. Additional page U R Ls are optional and use

9
00:00:43,783 --> 00:00:49,491
your plan allowance. A scan checks the initially loaded page, not every linked page or a signed-in journey.

10
00:00:49,491 --> 00:00:55,198
Wait for the result. If execution is incomplete, inspect the limitation and retry; do not treat it as

11
00:00:55,198 --> 00:01:01,906
a pass.

12
00:01:01,906 --> 00:01:08,811
Our first completed scan failed. It found one critical button-name violation affecting one element, with zero scan errors.

13
00:01:08,811 --> 00:01:15,716
Open the finding and read why it matters. A screen-reader user needs to know what the button does.

14
00:01:15,716 --> 00:01:23,621
A failed accessibility gate is useful evidence, not a reason to hide the result.

15
00:01:23,621 --> 00:01:30,831
Choose View locations and scan evidence. Here the selector is checkout, and the captured button has no text.

16
00:01:30,831 --> 00:01:38,041
Send the rule, location, observed page state and proposed fix to the developer. For this example, visible Checkout

17
00:01:38,041 --> 00:01:46,251
text is appropriate. Always verify the meaning of a label in the actual page.

18
00:01:46,251 --> 00:01:52,903
Find the result under My reports. Rename it with a useful baseline label, then choose Open report. Guests

19
00:01:52,903 --> 00:01:59,556
can use Save report and complete sign-in; that sign-in step is not shown here. Check that the intended

20
00:01:59,556 --> 00:02:07,208
report appears before leaving the page. Saved reports have an expiry, so download evidence you need to retain.

21
00:02:07,208 --> 00:02:14,139
In Saved tester review, select the finding or manual check. Set a tracking status, enter a reviewer name,

22
00:02:14,139 --> 00:02:21,070
and describe your test steps and evidence. Choose Save review. We kept this finding open. Our first keyboard

23
00:02:21,070 --> 00:02:29,002
attempt did not produce the expected message, and the note says so. Screen-reader testing remains outstanding.

24
00:02:29,002 --> 00:02:34,494
Choose Open reviewed report slash save P D F to include saved notes. It opens separately, so your

25
00:02:34,494 --> 00:02:39,986
working page stays available. Use its Print slash save as P D F control and your browser print

26
00:02:39,986 --> 00:02:45,478
dialog. Original downloads remain unchanged. The reviewed J S O N and review C S V are separate

27
00:02:45,478 --> 00:02:51,971
ways to retain your notes.

28
00:02:51,971 --> 00:03:00,513
The developer added visible Checkout text and corrected the demonstration script loading. On the updated page, the button

29
00:03:00,513 --> 00:03:09,055
now receives focus and Enter displays the no-payment confirmation. That is one observed interaction, not a complete manual

30
00:03:09,055 --> 00:03:18,597
assessment. Check keyboard order, visible focus, screen-reader behavior, zoom and relevant page states for your own application.

31
00:03:18,597 --> 00:03:24,030
Return to My reports and choose Rescan on the baseline. Review the U R L and confirm permission

32
00:03:24,030 --> 00:03:29,464
again. Our fresh scan of the same page passed. Rename it as the after report. A pass means

33
00:03:29,464 --> 00:03:34,897
this automated gate found no blocking result in that scan. It does not establish full W C A

34
00:03:34,897 --> 00:03:41,331
G conformance.

35
00:03:41,331 --> 00:03:48,236
In Compare reports, select the earlier baseline and the later retest. Choose Compare findings. Our comparison shows one

36
00:03:48,236 --> 00:03:55,141
finding no longer detected, zero newly detected, and zero still detected at the same location. Expand the result

37
00:03:55,141 --> 00:04:03,046
and verify it manually before closing the issue. Preserve both reports and your review evidence.

38
00:04:03,046 --> 00:04:11,866
Use the same loop on each authorized page: scan, inspect evidence, record manual observations, make a fix, and

39
00:04:11,866 --> 00:04:20,687
retest the same scope. Keep incomplete results and unreviewed checks visible. The developer guide covers local commands and

40
00:04:20,687 --> 00:04:30,508
GitHub Actions. Visit the testing guide on gatekeeper Q A dot com for the written workflow.